From a0788ec48cc3f1357075e8351aa7d62d20c63c15 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Thu, 25 Sep 2003 20:48:59 +0000 Subject: Update. * posix/regcomp.c (peek_token): Remove recent changes for anchor handling again. (parse_reg_exp): Likewise. * posix/regex.h: Remove RE_CARET_ANCHORS_HERE. (peek_token): Accept \s and \S as OP_SPACE and OP_NOTSPACE. (parse_expression): Replace build_word_op with --- ChangeLog | 9 +++++++-- posix/regcomp.c | 11 ++++++----- posix/regex.h | 5 ----- 3 files changed, 13 insertions(+), 12 deletions(-) diff --git a/ChangeLog b/ChangeLog index 36bda970f2..f1619f9a90 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,5 +1,10 @@ 2003-09-25 Ulrich Drepper + * posix/regcomp.c (peek_token): Remove recent changes for anchor + handling again. + (parse_reg_exp): Likewise. + * posix/regex.h: Remove RE_CARET_ANCHORS_HERE. + * csu/Makefile ($(objpfx)version-info.h): Fix sed script to handle banners. @@ -239,8 +244,8 @@ * posix/regcomp.c (build_word_op): Rename like... (build_charclass_op): ...this. Accept two extra parameters, CLASS_NAME and EXTRA. Add EXTRA to the result, not only _. - (peek_token): accept \s and \S as OP_SPACE and OP_NOTSPACE. - (parse_expression): replace build_word_op with + (peek_token): Accept \s and \S as OP_SPACE and OP_NOTSPACE. + (parse_expression): Replace build_word_op with build_charclass_op, add new arguments, accept OP_SPACE and OP_NOTSPACE. * posix/regex_internal.h (re_token_type_t): Add OP_SPACE diff --git a/posix/regcomp.c b/posix/regcomp.c index 3d54f99363..e2f01fc499 100644 --- a/posix/regcomp.c +++ b/posix/regcomp.c @@ -119,7 +119,7 @@ static reg_errcode_t build_charclass (RE_TRANSLATE_TYPE trans, reg_syntax_t syntax); #endif /* not RE_ENABLE_I18N */ static bin_tree_t *build_charclass_op (re_dfa_t *dfa, RE_TRANSLATE_TYPE trans, - const unsigned char *class_name, + const unsigned char *class_name, const unsigned char *extra, int not, reg_errcode_t *err); static void free_bin_tree (bin_tree_t *tree); @@ -1660,11 +1660,12 @@ peek_token (token, input, syntax) token->type = OP_PERIOD; break; case '^': - if (!(syntax & (RE_CONTEXT_INDEP_ANCHORS | RE_CARET_ANCHORS_HERE)) && + if (!(syntax & RE_CONTEXT_INDEP_ANCHORS) && re_string_cur_idx (input) != 0) { char prev = re_string_peek_byte (input, -1); - if (!(syntax & RE_NEWLINE_ALT) || prev != '\n') + if (prev != '|' && prev != '(' && + (!(syntax & RE_NEWLINE_ALT) || prev != '\n')) break; } token->type = ANCHOR; @@ -1799,7 +1800,7 @@ parse (regexp, preg, syntax, err) bin_tree_t *tree, *eor, *root; re_token_t current_token; int new_idx; - current_token = fetch_token (regexp, syntax | RE_CARET_ANCHORS_HERE); + current_token = fetch_token (regexp, syntax); tree = parse_reg_exp (regexp, preg, ¤t_token, syntax, 0, err); if (BE (*err != REG_NOERROR && tree == NULL, 0)) return NULL; @@ -1846,7 +1847,7 @@ parse_reg_exp (regexp, preg, token, syntax, nest, err) { re_token_t alt_token = *token; new_idx = re_dfa_add_node (dfa, alt_token, 0); - *token = fetch_token (regexp, syntax | RE_CARET_ANCHORS_HERE); + *token = fetch_token (regexp, syntax); if (token->type != OP_ALT && token->type != END_OF_RE && (nest == 0 || token->type != OP_CLOSE_SUBEXP)) { diff --git a/posix/regex.h b/posix/regex.h index 1ab3e243a7..e251962cef 100644 --- a/posix/regex.h +++ b/posix/regex.h @@ -170,11 +170,6 @@ typedef unsigned long int reg_syntax_t; If not set, then case is significant. */ #define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1) -/* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only - for ^, because it is difficult to scan the regex backwards to find - whether ^ should be special. */ -#define RE_CARET_ANCHORS_HERE (RE_ICASE << 1) - /* This global variable defines the particular regexp syntax to use (for some interfaces). When a regexp is compiled, the syntax used is stored in the pattern buffer, so changing this does not affect -- cgit v1.2.3